ArabicNewSplits5_FineTuningAraBERT_run2_AugV5_k2_task5_organization

This model is a fine-tuned version of aubmindlab/bert-base-arabertv02 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9769
• Qwk: 0.6315
• Mse: 0.9769
• Rmse: 0.9884

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Qwk	Mse	Rmse
No log	0.1538	2	2.3311	-0.0013	2.3311	1.5268
No log	0.3077	4	1.6628	0.1590	1.6628	1.2895
No log	0.4615	6	1.4311	0.1387	1.4311	1.1963
No log	0.6154	8	1.2441	0.2061	1.2441	1.1154
No log	0.7692	10	1.2247	0.2173	1.2247	1.1066
No log	0.9231	12	1.2165	0.2366	1.2165	1.1030
No log	1.0769	14	1.2113	0.2113	1.2113	1.1006
No log	1.2308	16	1.1667	0.2226	1.1667	1.0801
No log	1.3846	18	1.1556	0.3171	1.1556	1.0750
No log	1.5385	20	1.1305	0.3646	1.1305	1.0633
No log	1.6923	22	1.1652	0.3534	1.1652	1.0794
No log	1.8462	24	1.0784	0.4391	1.0784	1.0385
No log	2.0	26	1.0179	0.4527	1.0179	1.0089
No log	2.1538	28	0.9771	0.4712	0.9771	0.9885
No log	2.3077	30	0.9779	0.5319	0.9779	0.9889
No log	2.4615	32	0.9418	0.5666	0.9418	0.9704
No log	2.6154	34	0.9247	0.5707	0.9247	0.9616
No log	2.7692	36	0.9513	0.5725	0.9513	0.9754
No log	2.9231	38	0.9385	0.6034	0.9385	0.9688
No log	3.0769	40	0.8972	0.6130	0.8972	0.9472
No log	3.2308	42	0.9195	0.6223	0.9195	0.9589
No log	3.3846	44	1.0167	0.6120	1.0167	1.0083
No log	3.5385	46	1.0966	0.5788	1.0966	1.0472
No log	3.6923	48	0.9813	0.6009	0.9813	0.9906
No log	3.8462	50	0.8969	0.6426	0.8969	0.9470
No log	4.0	52	0.8435	0.6671	0.8435	0.9184
No log	4.1538	54	0.8375	0.6681	0.8375	0.9152
No log	4.3077	56	0.8864	0.6457	0.8864	0.9415
No log	4.4615	58	0.8963	0.6282	0.8963	0.9467
No log	4.6154	60	0.8959	0.6198	0.8959	0.9465
No log	4.7692	62	0.8998	0.6198	0.8998	0.9486
No log	4.9231	64	0.9578	0.6222	0.9578	0.9787
No log	5.0769	66	1.0791	0.6150	1.0791	1.0388
No log	5.2308	68	1.0082	0.6177	1.0082	1.0041
No log	5.3846	70	0.8780	0.6267	0.8780	0.9370
No log	5.5385	72	0.8552	0.6461	0.8552	0.9248
No log	5.6923	74	0.8676	0.6522	0.8676	0.9315
No log	5.8462	76	0.9809	0.5979	0.9809	0.9904
No log	6.0	78	1.2209	0.6274	1.2209	1.1050
No log	6.1538	80	1.3290	0.6091	1.3290	1.1528
No log	6.3077	82	1.2102	0.6291	1.2102	1.1001
No log	6.4615	84	1.0097	0.6245	1.0097	1.0048
No log	6.6154	86	0.8724	0.6629	0.8724	0.9340
No log	6.7692	88	0.8568	0.6653	0.8568	0.9256
No log	6.9231	90	0.8711	0.6543	0.8711	0.9333
No log	7.0769	92	0.9204	0.6368	0.9204	0.9594
No log	7.2308	94	1.0168	0.6159	1.0168	1.0083
No log	7.3846	96	1.0432	0.6273	1.0432	1.0214
No log	7.5385	98	1.0170	0.6146	1.0170	1.0085
No log	7.6923	100	1.0023	0.6159	1.0023	1.0012
No log	7.8462	102	0.9644	0.6192	0.9644	0.9820
No log	8.0	104	0.9007	0.6483	0.9007	0.9491
No log	8.1538	106	0.8847	0.6328	0.8847	0.9406
No log	8.3077	108	0.8727	0.6414	0.8727	0.9342
No log	8.4615	110	0.8836	0.6284	0.8836	0.9400
No log	8.6154	112	0.9079	0.6354	0.9079	0.9528
No log	8.7692	114	0.9594	0.6239	0.9594	0.9795
No log	8.9231	116	1.0166	0.6187	1.0166	1.0083
No log	9.0769	118	1.0287	0.6179	1.0287	1.0142
No log	9.2308	120	1.0190	0.6205	1.0190	1.0095
No log	9.3846	122	0.9986	0.6255	0.9986	0.9993
No log	9.5385	124	0.9775	0.6315	0.9775	0.9887
No log	9.6923	126	0.9756	0.6315	0.9756	0.9877
No log	9.8462	128	0.9783	0.6315	0.9783	0.9891
No log	10.0	130	0.9769	0.6315	0.9769	0.9884

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1